A few years ago Destroyer X (with Hugh's permission) published the conceptual schematic of one of Hugh's commercial AKSA 55 kitset products.
Here is the thread for reference:
http://www.diyaudio.com/forums/ever...ugh-thinking-let-us-see-schematics-forum.html
I was learning EAGLE at the time so I drew the schematic and designed a PCB as an exercise. At the time the AKSA 55 was still a commercial product, so I didn't think it was appropriate to publish the PCB. But now, the AKSA 55 is no longer for sale, so I sent the PCB files off to a PCB manufacturer and have now built a few of these amplifiers.
My schematic is a little different from the one that was shown in the reference thread in a couple of ways. I changed a couple of values that were obviously wrong, I used BD139 / BD140 as drivers and I used a BD139 for the VAS. When testing the PCBs I used the recommended transistors mentioned in the quote below.
I have actually built one amplifier using BD139 / BD140 transistors and will report on that at a latter date.
Here is the thread for reference:
http://www.diyaudio.com/forums/ever...ugh-thinking-let-us-see-schematics-forum.html
I was learning EAGLE at the time so I drew the schematic and designed a PCB as an exercise. At the time the AKSA 55 was still a commercial product, so I didn't think it was appropriate to publish the PCB. But now, the AKSA 55 is no longer for sale, so I sent the PCB files off to a PCB manufacturer and have now built a few of these amplifiers.
My schematic is a little different from the one that was shown in the reference thread in a couple of ways. I changed a couple of values that were obviously wrong, I used BD139 / BD140 as drivers and I used a BD139 for the VAS. When testing the PCBs I used the recommended transistors mentioned in the quote below.
I have actually built one amplifier using BD139 / BD140 transistors and will report on that at a latter date.
Last edited:
Some say, keeping things simple is a bless, tell me Greg, how does it sound? I am dying to know. Are you going to tweak the values to obtain the best performance?
I like this Zobel module, very attractive indeed. Output terminal is in the perfect location, placement of Q4 is also perfect to me. The row of resistors gives it a nice touch. Well done Greg.
I like this Zobel module, very attractive indeed. Output terminal is in the perfect location, placement of Q4 is also perfect to me. The row of resistors gives it a nice touch. Well done Greg.
Last edited:
Greg,
My thanks for keeping the old AKSA alive.
Metal,
This amp is utterly conventional in its engineering. It has simply been tweaked down to the last hertz to extract the best possible sound for the human ear. It wouldn't be much good for bats, for example, or Martians, and the magic sound qualities are purely dimensioning and component choice.
It sold by the hundreds around the world, quite a successful amplifier. Worth a good listen.
Hugh
My thanks for keeping the old AKSA alive.
Metal,
This amp is utterly conventional in its engineering. It has simply been tweaked down to the last hertz to extract the best possible sound for the human ear. It wouldn't be much good for bats, for example, or Martians, and the magic sound qualities are purely dimensioning and component choice.
It sold by the hundreds around the world, quite a successful amplifier. Worth a good listen.
Hugh
I have had a few inquires, so....
I should emphasize that the schematic I have published is based on the "conceptual" schematic Carlos published in his thread (see reference in post #1). It is not identical to any of the production AKSA 55's I am aware of. This means, it will not sound exactly like Hugh's production AKSA 55.
regards
I should emphasize that the schematic I have published is based on the "conceptual" schematic Carlos published in his thread (see reference in post #1). It is not identical to any of the production AKSA 55's I am aware of. This means, it will not sound exactly like Hugh's production AKSA 55.
regards
Ksc3503
To early to comment on the sound, it is only working on my work bench. I plan to test various tweaks to experience them for myself. So far, the amp on the left has BD139/BD140 for drivers and a BD139 for the VAS while the other 2 have the recommended drivers and a KSC3503 for the VAS.
I got sick of trying to squeeze the zobel and output inductor on the amp PCBs. This way I can built a half dozen of them suitable for testing the amps. This amp PCB has room for the zobel, but I am not using it. Thanks for the compliment.
I plan to build 2 mono blocks with 3 amps per heat sink. This way I can quickly compare 3 options in quick succession. I might automate the switching between them one day.
regards
To early to comment on the sound, it is only working on my work bench. I plan to test various tweaks to experience them for myself. So far, the amp on the left has BD139/BD140 for drivers and a BD139 for the VAS while the other 2 have the recommended drivers and a KSC3503 for the VAS.
I got sick of trying to squeeze the zobel and output inductor on the amp PCBs. This way I can built a half dozen of them suitable for testing the amps. This amp PCB has room for the zobel, but I am not using it. Thanks for the compliment.
I plan to build 2 mono blocks with 3 amps per heat sink. This way I can quickly compare 3 options in quick succession. I might automate the switching between them one day.
regards
Attachments
Last edited:
There are theoretical based experts, there are empirical based experts, that's if you can separate the two. I think Hugh started as an expert empirically (don't ask my words please) that's why he's so knowledgeable. It's different with those who started as theoretical experts and more or less remain there.
I think you can get good sound from simple circuits, but there are things that you have to do it specific in your implementation. Simulation program doesn't work here. Measuring equipment is better but listening test is also has the same function. I don't think you can easily copy the Aksa even if you have the complete schematic.
Reading this thread I thought you would finally update this schematic you posted long ago (because of the out-of-production issue you mentioned). But it turned out I'm wrong. As far as how I can understand it, many Aksa's ideas are not implemented in your schematic. Your schematic is very old and I don't think the schematic itself is the complete idea of this Aksa amplifier. If you follow all the available information about Aksa, I believe you can duplicate it because Hugh imo had explained all, tho not in the same place and not explicitly.
I think you can get good sound from simple circuits, but there are things that you have to do it specific in your implementation. Simulation program doesn't work here. Measuring equipment is better but listening test is also has the same function. I don't think you can easily copy the Aksa even if you have the complete schematic.
Reading this thread I thought you would finally update this schematic you posted long ago (because of the out-of-production issue you mentioned). But it turned out I'm wrong. As far as how I can understand it, many Aksa's ideas are not implemented in your schematic. Your schematic is very old and I don't think the schematic itself is the complete idea of this Aksa amplifier. If you follow all the available information about Aksa, I believe you can duplicate it because Hugh imo had explained all, tho not in the same place and not explicitly.
Member
Joined 2009
Paid Member
AKSA - the amplifier that has inspired and launched many a DIY attempt. It's what got me into this hobby. I started with the same schematic posted at the start of this thread for my TGM1 amplifier. It's good to see that the AKSA amplifier still has the same power to inspire.
I think Hugh knows it would be a disservice to reveal the secrets now, far better to keep the legend....
I think Hugh knows it would be a disservice to reveal the secrets now, far better to keep the legend....
Jay,
Apa kabar? Anda berasal dari Indonesia, atau negeri barat, mungkin AS? Saya di JKT Jan 2009, berkunjung keluarga istri di SMG.
You are essentially correct, there are some issues not covered. A quick email to Greg will clear most of them up, however. I do not wish to put this valuable circuit in full detail on a public forum, because I know some considerate fellow will start to make them in quantity, and since it's my IP I feel I should make it tough for him.
I agree with NP. A good amp is one third topology, one third layout, and one third parts choice. When it's all weighed up the topology is only a small part of it.
Cheers,
Hugh
Apa kabar? Anda berasal dari Indonesia, atau negeri barat, mungkin AS? Saya di JKT Jan 2009, berkunjung keluarga istri di SMG.
You are essentially correct, there are some issues not covered. A quick email to Greg will clear most of them up, however. I do not wish to put this valuable circuit in full detail on a public forum, because I know some considerate fellow will start to make them in quantity, and since it's my IP I feel I should make it tough for him.
I agree with NP. A good amp is one third topology, one third layout, and one third parts choice. When it's all weighed up the topology is only a small part of it.
Cheers,
Hugh
Hi Hugh, I'm from Jakarta
I posted on this thread because I was surprise about Aksa55 being not in production anymore and the (what I thought was Greg's) following statements regarding the possibility to expose those which was not possible before.
I tried to re-read the original post but I couldn't find the statement, so okay may be it's my hallucination
Just after I posted my comment, it came to my thinking that it is of course impossible that Greg doesn't know your circuit. And I thought I knew what is going on and I will shut up.
Yes, I think you and NP are alike in term of empirical expertise. Parts choice is indeed important especially with mosfets. And it was a surprise to me that it takes so long for NP to come up with Aleph J (It seems impossible that he just needed to get rid a ton of his IRF9610 in his warehouse). That's why I think I need to start studying to design my own amp. Because it is impossible to wait for him to design amplifier that I already "have" in my mind.
I'm surprised at that tbh
Topology is everything when it comes to sonics... layout and wiring is of absolute utmost importance... and parts ? well IMO as long as they are good quality commercial grade then that's the least important factor. Choose good parts for long life/performance, but I would never pay for "boutique" stuff.
Topology is everything when it comes to sonics... layout and wiring is of absolute utmost importance... and parts ? well IMO as long as they are good quality commercial grade then that's the least important factor. Choose good parts for long life/performance, but I would never pay for "boutique" stuff.
Greg is also the owner of Hugh's AKSA 55 and had upgraded to the 55N+ so will have Hugh's original design to compare the Baby AKSA.
I'd be very surprised if it doesn't have a similar presentation to Hugh's commercial AKSA.
I haven't built the Baby AKSA but have built several of Hugh's and have found the cap used at C4 can have a significant bearing on the final sound. I had best results with a Nichicon KZ (100uF 50V) in that location.... hint don't use a BG standard. Nichicon FG worked well for the other electrolytics but I don't think would be better than any other good quality cap of similar quality.
I'd be very surprised if it doesn't have a similar presentation to Hugh's commercial AKSA.
I haven't built the Baby AKSA but have built several of Hugh's and have found the cap used at C4 can have a significant bearing on the final sound. I had best results with a Nichicon KZ (100uF 50V) in that location.... hint don't use a BG standard. Nichicon FG worked well for the other electrolytics but I don't think would be better than any other good quality cap of similar quality.
Yes I do have an AKSA or two. I've also got a DIGI 125 that is similar. I am familiar with circuits and component values and I know rabbitz's mods as well.
I was definitely going to bring this up in a week or two. A Nichicon KZ in C4 is a very good recommendation, 95 cents well spent. BTW, the 100uF Nichicon KZ (50V) and FG (63V) are the same price. Peter, would the KZ's work as well as the FG's in the non-bootstrap locations. I would work out a little cheaper for quantity discounts if I got one type.
regards
I'm surprised at that tbh
Topology is everything when it comes to sonics... layout and wiring is of absolute utmost importance... and parts ? well IMO as long as they are good quality commercial grade then that's the least important factor. Choose good parts for long life/performance, but I would never pay for "boutique" stuff.
Hi Mooly,
I have a foot in each camp. I have personally experienced the differences a few components can make to simple circuit like this, but I also have friends with more politically correct amps using almost generic parts that sound wonderful as well. It might be the ear of the designer/tuner that is most important.
regards
Mooly, Luke
This is a subjective judgement.... and you can't be objective about it. It's just opinion, and should be taken with a crystal of chloride of sodium.
I prefaced my assessment with my agreement with NP, hoping it would make a cast iron point, but then, I'm not NP, am I? No one argues with him, to my great pleasure I might add. He has a proven track record after all.....
Hugh
This is a subjective judgement.... and you can't be objective about it. It's just opinion, and should be taken with a crystal of chloride of sodium.
I prefaced my assessment with my agreement with NP, hoping it would make a cast iron point, but then, I'm not NP, am I? No one argues with him, to my great pleasure I might add. He has a proven track record after all.....
Hugh
If i were to design a +50kn high speed vessel i'd have to choose the hull type (topology), the type of propulsion and materials used (parts), and tweak the hull geometry with model tank testing (layout).
Each of those components has an equal share in the final outcome.
Even the best commercial grade steel is useless compared to sealium alloy or carbon/aramid fiber composites.
Even the lowest kg/kW modern diesel engine is inferior compared to a 0.2kg/kW Vericor marine gas turbine.
Next parts step, above 30 knots speed nothing outperforms efficiency numbers of water jet propulsion systems.
The chosen hull topology often inherently determines the choice of parts.
Picking good quality general commercial parts may get you to the finish line more cost efficiently, but you will not have pole position and not finish first either.
Dismissing any part that is not generic commercial grade by labelling it as "boutique" is the equivalent to deciding not to compete for the design performance target from the onset.
About as pointless as the opposite approach : hoping to finish 1st with unsubstantiated, no one does voodoo like you do, boutique parlor items solely.
Member
Joined 2009
Paid Member
I've built my 'clone' of AKSA with the TGM1 amp and I can testify to the importance of part choices and the subtle but important effect of optimizing the values of the compensation components.
I haven't tried different layouts & wiring, but it doesn't take much effort to see the impact of parasitic capacitance and inductance using Spice. It's certainly important.
And don't forget the power supply, another essential element of the amplifier.
Anyhow, I hope everyone will respect Hugh's wishes to keep the details of his AKSA out of the public eye.
I haven't tried different layouts & wiring, but it doesn't take much effort to see the impact of parasitic capacitance and inductance using Spice. It's certainly important.
And don't forget the power supply, another essential element of the amplifier.
Anyhow, I hope everyone will respect Hugh's wishes to keep the details of his AKSA out of the public eye.